| Time |
Nick |
Message |
| 05:01 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 07:09 |
|
rjackson_isl joined #evergreen |
| 07:27 |
|
rlefaive joined #evergreen |
| 07:31 |
|
JBoyer joined #evergreen |
| 07:32 |
|
agoben joined #evergreen |
| 08:19 |
|
collum joined #evergreen |
| 08:53 |
|
mmorgan joined #evergreen |
| 08:57 |
|
bos20k joined #evergreen |
| 09:08 |
|
dbwells joined #evergreen |
| 09:09 |
|
Dyrcona joined #evergreen |
| 09:25 |
|
yboston joined #evergreen |
| 09:29 |
|
maryj joined #evergreen |
| 09:31 |
|
kmlussier joined #evergreen |
| 09:40 |
|
rlefaive joined #evergreen |
| 09:47 |
|
Freddy_Enrique joined #evergreen |
| 09:47 |
Freddy_Enrique |
Good morning everyone |
| 09:48 |
bshum |
Morning Freddy_Enrique! |
| 09:49 |
Dyrcona |
Good morning! |
| 09:49 |
kmlussier |
Good morning! |
| 09:50 |
kmlussier |
@coffee |
| 09:50 |
* pinesol_green |
brews and pours a cup of Colombia Hato Viejo Cup of Excellence, and sends it sliding down the bar to kmlussier |
| 09:51 |
Freddy_Enrique |
Colombian Coffee :) |
| 09:53 |
Freddy_Enrique |
Yesterday, I tried to install the evergreen but faild a couple of times. I could make these conclusions: Where it says user linux should be replaced with root, because it worked for me |
| 09:56 |
Dyrcona |
Actually, if you read carefully, I think it says to use the Linux root user. |
| 09:59 |
Freddy_Enrique |
let see, 5. Configuration and compilation instructions: with user linux |
| 10:00 |
Freddy_Enrique |
didnt accomplish anything with that user, I switched to root and then it worked :) |
| 10:00 |
Freddy_Enrique |
the same happens when trying to install opensrf |
| 10:01 |
Freddy_Enrique |
5. Configuration and compilation instructions (opensrf) Also points to use user linux |
| 10:01 |
kmlussier |
Freddy_Enrique: I've found that if I'm carefuly about switching users the way the documentation describes, the installation will work. The times the installation failed, it was because I had mistakenly done something under the wrong user. |
| 10:02 |
kmlussier |
But that step should work if you perform it under whichever user account you used to log in. I usually use the opensrf account as the user account. |
| 10:03 |
Freddy_Enrique |
Well, I tried to follow carefully the instruction. The only problem was with the user linux part, the rest was the same and it worked |
| 10:04 |
Freddy_Enrique |
I only used opensrf when the instructions indicated it |
| 10:04 |
Dyrcona |
Freddy_Enrique: They mean whatever user you are logged in as, not as user Linux. |
| 10:05 |
Dyrcona |
I usually create the opensrf account first, and do everything as that user, and sudo when it says to do something as root. |
| 10:05 |
|
abowling joined #evergreen |
| 10:05 |
Freddy_Enrique |
oh.... |
| 10:05 |
Freddy_Enrique |
by any chance Dyrcona, do you happened to use visudo? |
| 10:06 |
Freddy_Enrique |
since Im using ubuntu server, I needed to use that command |
| 10:06 |
Freddy_Enrique |
to give opensrf the permissions |
| 10:06 |
Dyrcona |
I do when I want to edit sudoers, but I usually add a file to /etc/sudoers.d. |
| 10:07 |
Freddy_Enrique |
its another way to give permissions right? |
| 10:08 |
Dyrcona |
Freddy_Enrique: you can also do 'usermod -a -G sudo opensrf' as root. |
| 10:08 |
Freddy_Enrique |
uhm, so weird. I was using my normal user and didnt successed, Gonna try again |
| 10:08 |
Dyrcona |
Freddy_Enrique: I create the opensrf user at O/S install time if I'm using the (virtual) machine only for Evergreen. |
| 10:09 |
Dyrcona |
I then login as opensrf, and it is usually the only user I create on the machine. |
| 10:09 |
Dyrcona |
If you do that with Ubuntu, the user is automatically added to the sudo group. |
| 10:10 |
Freddy_Enrique |
then...I can skip the create opensrf user? |
| 10:11 |
Freddy_Enrique |
this part: useradd -m -s /bin/bash opensrf |
| 10:11 |
Dyrcona |
Yes, if you've already created the user, you can skip that part. |
| 10:12 |
Freddy_Enrique |
Dyrcona++ |
| 10:12 |
Dyrcona |
I find that creating the user during installation, makes everything else easier. |
| 10:12 |
Dyrcona |
extraneous_comma-- :) |
| 10:32 |
|
Christineb joined #evergreen |
| 11:06 |
|
collum joined #evergreen |
| 11:45 |
|
Freddy_Enrique joined #evergreen |
| 11:48 |
|
RBecker joined #evergreen |
| 12:29 |
|
jihpringle joined #evergreen |
| 12:46 |
|
b_bonner joined #evergreen |
| 13:17 |
Dyrcona |
jeff Thought you might get a chuckle from this: 61 files changed, 95455 deletions(-) |
| 13:19 |
jonadab |
Nice. |
| 13:19 |
jonadab |
deletions++ |
| 13:19 |
Dyrcona |
They aren't completely gone, just moved to a different repo, so deleted from this one. |
| 13:19 |
jonadab |
Ah. I was thinking more deletions than additions usually means the code got easier to maintain. |
| 13:20 |
Dyrcona |
Well, it did, since these files are separate from the others. |
| 13:20 |
Dyrcona |
Most of these are one shots, too. |
| 13:43 |
jeff |
has anyone here written tests (pgtap or other) for circ policies? |
| 13:44 |
jeff |
"if i check this item out to this patron it should go out for this long / with this rule? even if you were just testing circ matrix matchpoints exclusively to start... |
| 13:45 |
Dyrcona |
No, but I have used the database functions to check that certain rules are working as expected. |
| 13:47 |
jeff |
Dyrcona: for your syrup install(s), are staff logging into syrup with their evergreen credentials, or are you setting a password for each syrup-using staff member in syrup itself? |
| 13:47 |
Dyrcona |
jeff: I think they're using their evergreen credentials, but I wasn't here for the initial setup. |
| 13:47 |
JBoyer |
I don't know if I understand databases anymore. I take a query with 7 inner joins and 12 outer, add *9 more* outer joins and exec time goes from 184 seconds to 41. Consistently (+- a couple seconds; it's not caching that helps). |
| 13:48 |
Dyrcona |
JBoyer: The magic of indexes.... :) |
| 13:48 |
JBoyer |
jeff, I haven't written circ policy tests, but it seems like it would take significant setup to actually test properly. |
| 13:49 |
jeff |
JBoyer: could be that one or all of the additional joins are forcing the planner to make a decision it wouldn't normally make -- and it ends up being a better decision after all! |
| 13:49 |
JBoyer |
I suppose I should throw a couple EXPLAINs at it, that should make for some interesting reading... |
| 13:50 |
JBoyer |
jeff, Oh, re: policy testing, are you thinking just some tests that you can run against your own real data to make sure things work the way you expect? |
| 13:50 |
jeff |
JBoyer: right. |
| 13:50 |
Dyrcona |
I'm looking through some of my old stuff to see what I've got. |
| 13:51 |
JBoyer |
That does sound interesting. |
| 13:51 |
jeff |
JBoyer: we used to do similar with ipchains firewall rules, though that became a non-option with the move to iptables / conntrack. It may have recently become possible again... |
| 13:51 |
Dyrcona |
I'd usually just spot check things when someone reported a problem. |
| 13:51 |
JBoyer |
You could turn that into something Nagios/Icinga/etc. could verify once a day too... |
| 13:51 |
Dyrcona |
And, you'd have to keep it up to date as policy changes are made..... :) |
| 13:52 |
jeff |
right. |
| 13:52 |
jeff |
that's where it could either fall down or truly shine, depending on how many people change circ policies for which org units how often, etc. :-) |
| 13:52 |
JBoyer |
Dyrcona, Sure, but it has a built in reminder to update the check to reflect the new policy. ;) |
| 13:52 |
Dyrcona |
heh |
| 13:53 |
Dyrcona |
IIRC, I mostly ran stuff in the database with the find_circ_matrix_matchpoint and related functions. |
| 13:56 |
jeff |
and for the exhaustive test, you just join actor.usr with asset.copy and actor.org_unit... |
| 13:59 |
Dyrcona |
heh |
| 13:59 |
Dyrcona |
Well, you basically need some representative transactions where you know what the outcome should be ahead of time....but you knew that. |
| 13:59 |
* jeff |
nods |
| 14:00 |
Dyrcona |
And, I'm not finding much, though I did chuckle about the sequence in the names of some scripts that were aimed resolving an issue.... |
| 14:00 |
Dyrcona |
ticekt#_fix_problem.sql |
| 14:00 |
jeff |
though you could compare two "comprehensive" runs and then report on ones that changed. |
| 14:01 |
Dyrcona |
ticket#_poke_me_in_the_eye.sql |
| 14:01 |
Dyrcona |
ticket#_poke_me_again.sql |
| 14:01 |
Dyrcona |
ticket#_with_a_hot_poker.sql |
| 14:01 |
Dyrcona |
I guess it took 4 tries to get it right. :) |
| 14:02 |
jeff |
joke's on you. the real solution is stored as a comment on the original ticket in the old ticket system which has since been decommissioned. |
| 14:02 |
jeff |
"don't know why i kept trying to fix it with database changes, it turned out to be..." |
| 14:02 |
jeff |
:-) |
| 14:03 |
Dyrcona |
"...in a locked filing cabinet in a disused lavatory with a sign on the door saying, 'Beware of Panther?'" |
| 15:40 |
|
Jillianne joined #evergreen |
| 16:06 |
|
kmlussier joined #evergreen |
| 16:11 |
Dyrcona |
Anyone seen this from apache2ctl-websockets before: AH00534: apache2: Configuration error: No MPM loaded. |
| 16:12 |
Dyrcona |
Looking at /etc/apache2-websockets/mods-enabled, it appears that mpm prefork is configured. |
| 16:15 |
Dyrcona |
Nothing in the apache logs, either. |
| 16:24 |
Dyrcona |
Hmm... Deliberately loading it via apache2.conf leads to more errors. |
| 16:25 |
Dyrcona |
Oh. I think I may have copied the wrong apache2.conf for websockets. |
| 16:26 |
Dyrcona |
Looks like I got the Apache 2.2 version and this host is running apache 2.4. |
| 16:29 |
Dyrcona |
Yeap. |
| 16:31 |
pinesol_green |
News from qatests: Test Failure <http://testing.evergreen-ils.org/~live> |
| 16:58 |
|
frank_guel joined #evergreen |
| 16:59 |
|
frank_guel joined #evergreen |
| 17:03 |
frank_guel |
Hi all, I am still trying to upgrade from 2.8.4 to 2.12.3, (from one sorver to another too), all the migration is fine, but when I test creating a marc record something weird hapends, for example, If I create/import a new record with this "©" character, it is saved in the database with an strange character "�", but if I re-open the same marc record register, and replace the character with the correct one it is saved correc |
| 17:04 |
frank_guel |
I tested in two diferent server and is the same, |
| 17:06 |
|
mmorgan left #evergreen |
| 17:58 |
|
Freddy_Enrique joined #evergreen |
| 18:05 |
|
Freddy_Enrique joined #evergreen |
| 18:09 |
Freddy_Enrique |
One question guys, is it possible to enter the staff client with the same user from different computers? |
| 18:09 |
berick |
Freddy_Enrique: yes |
| 19:08 |
Freddy_Enrique |
If i create the opensrf user during VM installation, which part of the Create and set up the opensrf Unix user environment I should omit? |
| 19:09 |
Freddy_Enrique |
useradd -m -s /bin/bash opensrf - this one definitely |
| 19:09 |
Freddy_Enrique |
anything else |
| 22:57 |
|
RBecker joined #evergreen |
| 23:52 |
|
dbwells_ joined #evergreen |
